-
EleutherAI
- Berkeley CA
- @norabelrose
Stars
Language: Python
Sort by: Most stars
Long Range Arena for Benchmarking Efficient Transformers
Render probabilistic graphical models using matplotlib
RobustBench: a standardized adversarial robustness benchmark [NeurIPS'21 Benchmarks and Datasets Track]
This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents.
Fast Differentiable Sorting and Ranking
Code for Deep RL from Human Preferences [Christiano et al]. Plus a webapp for collecting human feedback
Locating and editing factual associations in GPT (NeurIPS 2022)
Interactive Composition Explorer: a debugger for compositional language model programs
TensorFlow implementation of On the Sentence Embeddings from Pre-trained Language Models (EMNLP 2020)
MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Research
Code release for "Git Re-Basin: Merging Models modulo Permutation Symmetries"
Repository for the paper "Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images"
Simple, Elegant, Typed Argument Parsing with argparse
Sparse learning library and sparse momentum resources.
Supplementary code for the paper "Hyperbolic Image Embeddings".
Newton and Quasi-Newton optimization with PyTorch
ADAHESSIAN: An Adaptive Second Order Optimizer for Machine Learning
alpha-beta-CROWN: An Efficient, Scalable and GPU Accelerated Neural Network Verifier (winner of VNN-COMP 2021, 2022, and 2023)
Erasing concepts from neural representations with provable guarantees
python tool to extract smoothed centerlines from polygons
PyTorch implementation of "Lagging Inference Networks and Posterior Collapse in Variational Autoencoders" (ICLR 2019)
Keeping language models honest by directly eliciting knowledge encoded in their activations.
Numerical integration in arbitrary dimensions on the GPU using PyTorch / TF / JAX
An extremely well tested and commented red black tree implementation. Worth a look if you are studying the material.